Investigate optimization of DB
It seems that many of the CZI records being imported don't have much data and exist mainly to act as counts for aggregation numbers like Source DOI count and Accession number count Is there a way we could avoid having millions and millions of assertion records since some operations take extremely long on data of this size